A Framework of Temporal Data Retrieval for Unreliable WSNs Using Distributed Fountain Codes

Distributed storage coding has been widely applied on data gathering over unreliable wireless sensor networks (WSNs), where it is essential to ensure the data persistence in case of several sensor failures caused by battery run-out or some physical damage problems surroundings. How to efficiently disseminate and collect the sensing data over WSNs is a key challenge yet. In this study, assumed that there are K sensor nodes equipped with sensing apparatus within N storage sensors, these K numbers of sensors can sense environmental changes and disseminate coded (by Fountain codes) time-series data over WSNs using the simple random walk. In order to perform the Fountain codes over WSNs, the question is to disseminate data in the long range of random walks to preserve the randomness so as to promote the source decoded rate. In this paper, a framework with partial decoding is proposed due to the temporal dependency of time-series sensing data. The reasons are twofold: (a) the complete decoding is not necessary for time-series data since the missing portions can be compensated by that of neighbors; (b) even if the ideal Luby transform (LT) code is optimized in terms of convergence, the complete decoding process is high power-consuming. Furthermore, a mathematical model to estimate the appropriate source decoded rate is given to guarantee the error tolerable level (<; 4% normalized root-mean-square error (NRMSE)). Experimental results show that the communication cost is affordable in the real cases.